NASA-CR-204434

Flood Management EnhancementUsing Remotely Sensed Data

Final Project Report

OJ _/cp _ "7

Prepared by

Gregory J. Romanowski

Sentar4910 Corporate Drive, Suite C

Huntsville, AL 35805

(205) 430-0860

May 1997

Cooperative Agreement No. NCC5-111

Prepared for

NASANational Aeronautics and Space Administration

Goddard Space Flight Center, Code 912

Greenbelt Road

Greenbelt, MD 20771

"IHE VIEWS, OPINIONS, AND/OR FINDINGS CONTAINED IN THIS REPORT ARE THOSE OF "IHE AlYrHOR(S)AND SHOULD NOT BE CONSTRUED AS AN OFFICIAL GOVERNMENT POSITION, POLICY, OR DECISIONUNLESS SO DESIGNATED BY OTHER OFFICIAL EX3CUMENTATION.

https://ntrs.nasa.gov/search.jsp?R=19970017778 2020-06-27T20:48:05+00:00Z

TABLE OF CONTENTS

Section Pa_e

INTRODUCTION ................................................................................................................ 1

Background ................................................................................................................. 1

Scope ........................................................................................................................... 1

Purpose ........................................................................................................................ 1

SUMMARY OF MILESTONE PERIOD ACTIVITIES ...................................................... 2

Significant Accomplishments ..................................................................................... 2

Prototype System Development Status .............................................................. 2

Software Module Development ......................................................................... 2

Remote Sensing Data Acquisition and Processing ............................................ 4

GIS Data Layer Development ............................................................................ 7

Prototype Demonstrations ................................................................................ 10

Scenario 1: Southeast Alabama Flood ............................................................ 11

Scenario 2: Northern Alabama Flood ............................................................. 12

Other Activities ......................................................................................................... 16

Home Page ....................................................................................................... 16

Principal Investigator Conference ................................................................... 16

Civil Air Patrol ................................................................................................. 16

Problems Encountered .............................................................................................. 16

Data Acquisition .............................................................................................. 16

Internet Connectivity ....................................................................................... 16

Clarification of Collaborative Agreements ...................................................... 16

User Identification ........................................................................................... 16

Milestones ................................................................................................. ....... 16

Requests for Assistance from NASA ........................................................................ 17

FUTURE ACTIVITIES ..................................................................................................... 17

APPENDICES

A. Project Summary .............................................................................................. A-1

B. CD Readme and Processing Description Files ................................................. B-1

C. Briefing Slides for the NASA/FEMA Conference on GIS and Applications of

Remote Sensing ................................................................................................ C-1

_ 97_2_

Figures Pa_e

1. Prototype System Concept .......................................................................................... 3

2. Sample Display from the EMO Tool Help Topics ..................................................... 4

3. Area Covered by the Processed Landsat TM Data ..................................................... 5

4. Example of the Map-X Simulated True Color Image Processing .............................. 6

5. Water Classification Image File Generated from Map-X ........................................... 8

6. Example of the USGS NAPP Photographs with TIGER GIS Data ............................ 8

7. Alabama Counties in the Prototype System ................................................................ 9

8. Sample of the TIGER Derived Data Coverages for Northern Alabama ..................... 9

9. In-land Wind Output for Scenario ............................................................................ 13

10. Sample of the Products Produced for Scenario 1 ..................................................... 13

11. Landsat TM Band 7 and TIGER Road, Hydrology, and Railroad Data ................... 14

12. Landsat TM and TIGER Data with Aerial Photograph ............................................ 15

13. Figure 12 with Landsat Derived Water Classification .............................................. 15

Tables Pa_e

1. Data Layers Provided for Prototype System ............................................................. 10

i 97-0269

FLOOD MANAGEMENT ENHANCEMENT

USING REMOTELY SENSED DATA

FINAL PROJECT REPORT

INTRODUCTION

Background. SENTAR, Inc., entered into a cooperative agreement with NASA Goddard Space

Flight Center (GSFC) in December 1994. The intent of the NASA Cooperative Agreement was

to stimulate broad public use, via the Internet, of the very large remote sensing databases

maintained by NASA and other agencies, thus stimulating U.S. economic growth, improving the

quality of life, and contributing to the implementation of a National Information Infrastructure.

SENTAR headed a team of collaborating organizations in meeting the goals of this project.

SENTAR's teammates were the NASA Marshall Space Flight Center (MSFC) Global Hydrology

and Climate Center (GHCC), the U.S. Army Space and Strategic Defense Command

(USASSDC), and the Alabama Emergency Management Agency (EMA).

For this cooperative agreement, SENTAR and its teammates accessed remotely sensed data in

the Distributed Active Archive Centers, and other available sources, for use in enhancing the

present capabilities for flood disaster management by the Alabama EMA. The project developed

a prototype software system for addressing prediction, warning, and damage assessment for

floods, though it currently focuses on assessment. The objectives of the prototype system were

to demonstrate the added value of remote sensing data for emergency management operations

during floods and the ability of the Internet to provide the primary communications medium for

the system. To help achieve these objectives, SENTAR developed an integrated interface for the

emergency operations staff to simplify acquiring and manipulating source data and data products

for use in generating new data products. The prototype system establishes a systems

infrastructure designed to expand to include future flood-related data and models or to include

other disasters with their associated remote sensing data requirements and distributed datasources.

Scope, This report covers the specific work performed during the seventh, and final, milestone

period of the project, which began on 1 October 1996 and ended on 31 January 1997. In

addition, it provides a summary of the entire project.

Purpose. The purpose of this report is to: 1) document the successful completion of all the

seventh period milestones and 2) to provide a status of the significant accomplishments,

problems encountered, and overall project results. The document presents the results of the

prototype demonstrations in the SUMMARY OF ACTIVITIES section. This is followed by a

FUTURE ACTIVITIES section. In addition, there is a summary of the entire project presented

in Appendix A. This information is intended to provide project insight to NASA and others in

applying NASA remote sensing data to disaster management.

1 97-0269

SUMMARY OF MILESTONE PERIOD ACTIVITIES

Significant Accomplishments. SENTAR had two products for the seventh milestone period.These were:

1. Completion of the final project report

2. Successful completion of the prototype demonstrations.

These two products have been successfully completed. This report constitutes the completion of

the first product. The second milestone was achieved with the completion of the prototype

demonstrations in December, with follow up interviews of the demonstration participants in

January and February. Details of the demonstrations are described in the following paragraphs.

Prototype System Development Status. The prototype system consists of four major

geographically-separated components. The major system components are a Data Processing

Center (DPC) (located in Huntsville, AL), the Alabama EMA Emergency Operations Center

(EOC) (in Clanton, AL), the remote sensing data providers (various locations in the world), and

the remote or county users (at various locations within the state of Alabama). For the prototype

system, the DPC is split between two facilities: the NASA GHCC for remote sensing data

processing and the USASSDC for the Geographic Information System (GIS) and ARC/INFO

conversion processing.

During the sixth milestone period, there were upgrades to the hardware and software at both

DPC facilities. In addition, Alabama EMA completed a major upgrade of their computer

resources. A new Sun server was installed and new, more capable PCs were provided for the

emergency management staff. These improvements were specifically provided to support the

Federal EMA's (FEMA's) Emergency Management Information System for the chemical

stockpile incineration program, however they are also supporting this effort. These replacement

PCs provide the computing and storage capacities necessary to run ArcView adequately with the

large data files that will be used. Prior to the prototype experimentation and demonstrations, the

remote sensing and GIS data file processing were completed and the files were installed on the

Sun server and some of the PCs. In addition, the revised version of the Emergency Management

Operations Tool (see following subsection for details) was loaded onto the PCs and configuredfor their local area network.

Software Module Development. The prototype system consists of a set of commercial-off-the-

shelf applications integrated under a common user interface, called the Emergency Management

Operations (EMO) Tool. The system makes extensive use of existing software products at the

Alabama EMA to provide system functionality wherever possible. Where these software

packages did not meet the functional needs of the prototype system, SENTAR developed

software applications and utilities. Three software components were developed for this project:

1. The EMO Tool

2. The Vector/Raster Data Formatter

3. Data Retrieval Scripts.

The EMO Tool resides on the PCs of the EOC operations center, the Vector/Raster Data

Formatter operates on PCs at the DPC, and the Data Retrieval Scripts operate on the Sun server

at the EOC. (Refer to Figure 1.) These three software components were developed and tested

"_ _ 97-0269

during previous milestone periods. During the seventh milestone period the only change to these

products was the addition of on-line help to the EMO Tool. This was set up like a standard

Windows help utility with topic selection from search or contents options. Figure 2 illustrates a

portion of the Windows help display that was added to the EMO Tool.

_p SPOT & _.[orier RSD /I

\_.2 ,nformaton 'b "-_- ._. RSO S_'_

.Photography_)"qi_... l_w_w._'--.._. / ,_irc:tafc I

Da la Queries. _'_ /& Requesl_ "-- ._

., , Spatial ._Jl_I_l Fealure _

S tiaIOa I.:.oufo@s

Istal_ Agendes

Dala Ptooes_ingCenter

_laba_ma

I_ A EOC ..... ,

Slate Sun Sorer

Spatial _o _sData n

O_er RSDSouro_s

/-SPOT, GOES,

TVA, N WS,Ne xRad, _. Olher

Wearier Silesi

//

/

_'/E-M'ai EM Informaton

E_ Data I Sources__

Selecled Warning,_sessmen_& I I

FIeoo_ery Images IGeneralPuBic& Text [

Rood I_edioton, earning, _R.a_e_n_ntinformation & images

FEMA • I.. ' "

Coun_ B,,,1Ofiee

Legend Ii --1 -= -- Ii

Ph_t_ioal Fax IMedia

Cellular InlernetComrnunicalons

I

Figure 1. Prototype system concept.

_ 97-0269

Icons

Goes Images

The Space Science and Engineering Center (SSEC) is a global leader in the analysis and distributionof geostationary satellite data. The SSEC is also a part of the University of Wisconsin-Madison's

Graduate School. SSEC is concerned mainly with scientific research and technology developmentthat enhances understanding of the atmosphere of Earth and also looking down to gain information

and insights into weather, climate, and other aspects of our global environment.

The SSEC provides real-time data for visible and infrared images from the (GOES) satellite. Thefollowing GOES images are:

North America GOES Visible (updated every 30 minutes)

North America GOES Infrared (updated every 30 minutes)

Doppler Images

Doppler Radar detects the presence and location of an object by bouncing an electromagnetic signal

off of it and measuring the time it takes for the signal to return. This measurement is used to

determine the distance and direction of the object from the radar. In the case of radar meteorology,the "objects" being measured are the particles of water, ice or dust in the atmosphere. Doppler radars

take additional advantage of the fact that radar signals reflected from a moving object undergo achange in frequency related to the speed of the object traveling to or away from the radar antenna.

Figure 2. Sample display from the EMO Tool help topics.

Two of the three software components, the EMO Tool and the Vector/Raster Data Formatter,

were installed and demonstrated on the appropriate platforms. The Data Retrieval Scripts were

not installed on the Alabama EMA server for the exercises/demonstrations, as direct Internet

access was not available for the demonstrations. Instead, these were run from SENTAR's ISP

computers and downloaded to the Alabama EMA via their dial up connection.

Remote Sensing Data Acquisition and Processing. During the seventh milestone period,

SENTAR finalized processing of the Landsat TM imagery that is the primary imagery backdrop

for the prototype system and produced the data on CDs. The satellite data cover all of northern

Alabama and the southeast portion of the state. (See Figure 3.) The data were reformatted for

importing into the Map-X remote sensing software where the individual bands were combined

into a single file and then converted into the BIL file format for exporting to ArcView. The

composite band file is a large file (-300 Mbytes) that is cumbersome for the emergency

managers to work with when not all six bands are needed. Therefore, SENTAR also produced

three-band composite files. For these, three of the Landsat TM bands were combined into a

97-0269

single-band, color composite file. A Map-X algorithm was applied to enhance the colors of the

composite file. This algorithm generates a simulated true color image by analyzing the range of

pixel values used in the bands, then spreads out the color values across the 256 bits of each pixel

of the composite file. Figure 4 shows an example of the effect of the simulated true color image

processing. The resulting file was also converted into a BIL format for exporting to ArcView.

The color composites were generated by assigning Landsat TM bands 4, 3, and 2 to the colors

red, green, and blue, respectively.

Figure 3. Area covered by the processed Landsat TM data.

Finally, water classification raster files were produced from the Landsat image using the tools

within Map-X. (See Figure 5.) A median filter was applied to the resulting image file to

eliminate solitary pixels identified as water. Although these may, in fact, be water, the desired

output is for significant bodies of water and not every small pond. The boundaries for the water

classified areas were then converted into a vector representation from the raster image. The

vector file was converted into CAD format (.dxf) for exporting to ARC/INFO. Once in

ARC/INFO, the vector file was again converted, this time to the ARC/INFO .e00 file format and

ArcView .shp file formats. It is these last formats that were exported to Alabama EMA along

with the original raster band images and the simulated true color image.

97-0269

Landsat TM image bands 4 (red), 3 (green), and 2 (blue)

Map-X Simulated True Color Image of above Landsat TM image

Figure 4. Example of the Map-X simulated true color image processing.

97-0269

Finally, water classification raster files were produced from the Landsat image using the tools

within Map-X. (See Figure 5.) A median filter was applied to the resulting image file to

eliminate solitary pixels identified as water. Although these may, in fact, be water, the desired

output is for significant bodies of water and not every small pond. The boundaries for the water

classified areas were then converted into a vector representation from the raster image. The

vector file was converted into CAD format (.dxf) for exporting to ARC/INFO. Once in

ARC/INFO, the vector file was again converted, this time to the ARC/INFO .e00 file format and

ArcView .shp file formats. It is these last formats that were exported to Alabama EMA along

with the original raster band images and the simulated true color image.

These various image and vector files were grouped by the Landsat TM image from which theywere derived and were bumed onto a writeable CD. A readme file was also added to describe

the contents of the CD and a Process.txt file was added to describe the processing performed on

the original data to produce the data sets on the CD. Appendix B presents a sample of a readme

file and the Readme file. This format for the files on CD represented a prototype of what

production CDs consist of.

Also during this milestone period, SENTAR purchased U.S. Geological Survey (USGS) National

Aerial Photography Program (NAPP) and National High-Altitude Photography (NHAP) program

photographs. These are hardcopy images that were then scanned into a digital format for

processing. Several black-and-white and color images were processed into BIL files for

exporting to ArcView. The photographs acquired cover areas of southern Alabama that have

experienced major floods in recent years. These images cover ground areas of 5x5 miles or 8x8

miles. To date, individual photographs have been digitized, georeferenced, and converted to BIL

format for export to Alabama EMA. Figure 6 shows a sample of two of these images with someTIGER GIS data.

GIS Data Layer Development. The development of the GIS data layers for the prototype system

were completed during this milestone period. SENTAR acquired the necessary source data to

support the data layers for the Alabama counties to be included in the prototype system. The

desired data layers previously defined with Alabama EMA for the prototype system were divided

into three categories: those that can be produced from the U.S. Census Bureau's TIGER data,

those that can be obtained from other federal agencies, and those that will be obtained from state

sources. TABLE 1 lists the data coverages that were produced for the prototype system from the

desired data coverages for these three categories.

SENTAR contracted out for new sets of these data layers using the most current version of

TIGER data. These data layers were supplemented with data layers that were provided by the

U.S. Army from the Integrated Disaster Planning Package (e.g., the Redstone Arsenal boundary).

TIGER data for the counties of northern Alabama and southeast Alabama were combined and

the composite vectors were processed to correct for proper edge matching at the county

boundaries, dangling vectors, and open polygons. The completed data coverages were

completed and provided to SENTAR last milestone period. This milestone period SENTAR

performed a quality check of the data and installed the data coverages onto the Alabama EMAserver. Figure 7 shows a state view with the northern and southeast counties of Alabama for

which TIGER data coverages were provided. Figure 8 illustrates the level of detail in the GIS

vector files by displaying a small portion of south Madison County in northern Alabama. This

figure presents six of the data coverages, featuring the hydrology and road coverages.

'7 97-0269

Figure 5. Portion of a water classification image file generated from Map-X.

\

Figure 6. Example of the USGS NAPP photographs with some TIGER GIS data.

97-0269

Figure 7. Alabama counties in the prototype system.

Figure 8.

Q 97-0269

The next category was other federal agencies. For this data layer, the newly digitized data of the

Flood Insurance Rate Maps (FIRM) were obtained. FEMA developed digital FIRMS for 19

counties in Alabama, which are part of the Digital Q3 Flood Data. These data were purchased

and received from the FEMA Map Center during the sixth milestone period. All 19 counties

were reprocessed to convert the projection datum to North American datum 1983 (NAD83)

versus the NAD27 datum used in the source data. This conversion was necessary for these data

to be compatible with the Landsat raster imagery, which was provided to Alabama EMA in the

Universal Transverse Mercator projection, NAD83 datum. This processing was performed at the

USASSDC facilities using ARC/INFO. This processing was essential because ArcView does not

have the ability to perform the projection and datum conversions as can be done usingARC/INFO.

For the state agency data, Alabama EMA sent out formal letters requesting the desired data and

these were followed up with telephone contacts by SENTAR personnel to discuss the details of

the proposed data transfers. Data were provided by the Alabama Power Commission and a

power plant data coverage was generated. No other state agencies provided data to be developed

into data coverages prior to the exercises/demonstrations, however some are currently

developing GIS data coverage of their data and have agreed in principle to provide these data toAlabama EMA when available.

TABLE 1

DATA LAYERS PROVIDED FOR PROTOTYPE SYSTEM

Data Source Data Layer

TIGER Hydrology, boundaries (local, county, state, military,

and international), urban areas, block group, voting

district, school district, demographics (census tract),

interstate highways, roads and streets, railroads,

airports, runways, shopping centers, trailer parks,

apartment complexes

Other Federal Agencies Flood Insurance Rate Maps

State Agencies Power plants

Prototype Demonstrations. An experiment/demonstration plan was developed during the last

milestone period. This plan was designed to conduct a series of two simulated flood events in

which the overall prototype system is demonstrated. The activities of the plan exercise the

developed and integrated components of the system to show overall functionality and examine

how the prototype system works with the existing plans and procedures at the Alabama EMA

EOC. Each experiment/demonstration focuses on one of the geographical areas of the GIS data

coverages developed for this project. In addition, each experiment/demonstration exercises

1 tr_ 97-0269

particular aspects of the prototype system. There was one experiment/demonstration involving a

hurricane that makes landfall on the Florida panhandle and heads north through the southeast

counties of Alabama, causing flooding along several rivers and streams. The other

experiment/demonstration consisted of flooding in northern Alabama resulting from severe and

prolonged spring thunderstorms. In both cases, Landsat TM images were used as the backdrop

imagery in the GIS for the normal, or non-flooded, view. Then for the visualization of the flood,

Landsat TM imagery was used as a water derived classification to locate and assess the flood

regions in one case and aerial photography was used in the other.

The experiments/demonstrations were developed in separable portions since the complete

demonstrations were designed to take several days working on a part-time basis. The schedule

had to be easily modifiable since Alabama EMA's actual mission always had priority over these

activities and, thus, there might have been interruptions in the execution of the experiments/demonstrations.

The plans for the experiments/demonstrations were developed during the sixth milestone period

except for details of the integration of Alabama EMA specific operating procedures, which were

to be added during the first part of the final milestone period. However, Alabama EMA was

only beginning the process of determining who in the EOC should run the GIS and remote

sensing applications and how to utilize the resulting data and data products in their mission.

Thus, they were not ready to provide the detail SENTAR was looking for in the

exercise/demonstration plan. These exercises/demonstrations were then changed into two parts.

The first part was a training program to produce hands-on usage of the tools and data by

operations staff at the EOC. This was done to make sure they really understood what data was

being provided to them, and to get them working with the data to start the process of determining

how to integrate these data products into their specific missions. The second part consisted of

SENTAR and NASA GHCC personnel executing the exercise/demonstration plan instead of the

EOC staff This resulted in the successful data processing and transmission steps of the plan, but

produced no data on the utilization of the data products in actual EOC operations.

A second limitation of the exercises/demonstrations was that very little of the state GIS data

were produced during the project. It is these data sets that represent the most important GIS

information to be used by the emergency managers during actual emergencies. Although many

of the data sets are in development within state agencies, only the power plant information was

available for the exercises/demonstrations. The lack of additional state data coverages only

impacted the evaluation of disaster impact and recovery utilization of state infrastructure.

Scenario 1: Southeast Alabama Flood. This first scenario simulated a hurricane hitting the

southeast portion of the state and causing flooding on certain rivers in Alabama that tend to be

very flood prone. The scenario began with the monitoring of the storm off shore through the

GOES imagery that was downloaded regularly and viewed through the image viewer. GOES has

the resolution to accurately monitor the progress of major storms such as hurricanes. Since an

actual hurricane was not threatening Alabama, the GOES imagery was used to track major cloudactivity through the southern portion of the state and to track hurricanes in other areas of the

hemisphere. Similarly, the Internet links to the National Weather Service (NWS) were regularly

checked to get storm forecasts, warnings, and watches.

After monitoring the imaginary hurricane for several hours, landfall was declared and a run of

the inland winds generated. It plotted a path of the hurricane through the southeast portion of

| | 974)269

the state. (See Figure 9.) An ArcView project file was then built with the available TIGER,

power plant, FEMA floodplain, and water classification (nonflooded condition) GIS data. In

addition, the Landsat TM and USGS aerial photography for this region are added to the projectfile.

Based on NWS flood warnings, radar imagery, and NWS river reports retrieved from the

Internet, it was determined that there was possible flooding on the Pea River near Elba. The

USGS aerial photographs that had been processed for that area were now used to access the

nonflooded region along the river to begin planning route into and out of the area using theTIGER road data.

Also, the data providers were contacted to determine if there was a satellite image that could be

used to ascertain the extent of the flood. An available Landsat image was used as if it were an

actually procured image of the flood. It was processed at the DPC (currently at the NASA

GHCC). There the image was processed into a BIL formatted file for all six bands, the simulated

true color image, and water extent. The latter was also produced into a vector file and converted

into the ArcView format. All the files, except the six-band file can be transmitted over the

Internet to Alabama EMA even though they only have a 28.8 kbps dial up Internet connection.

The simulated true color image would take about four hours to transmit, however the reliability

of the connection for that long is questionable. One technique that can be used is to not export

the entire image, since the Landsat image covers a wider area than is typically needed except for

very major events. In addition, a CD would also be made to provide all files in a format that is

easily transportable and easily stored for archival purposes.

With the new image imported into ArcView, the capabilities of the ArcView tool were then used

to generate outputs of the original view of the river as a raster image and the new view as a

vector representation using the water classification vector file from the new image. The FEMA

floodplain data was also used to show where the 100 and 500 flood zones were in the area of

concern. Figure 10 shows a sample of the products produced. Since the images were not of anactual flood, there were no areas where the water classification had encroached on homes and

businesses.

Scenario 2: Northern Alabama Flood. This second scenario simulated flooding on the

Tennessee River near Decatur, Alabama, resulting form severe and prolonged spring

thunderstorms. The scenario began with the monitoring of the storms through the radar imagery

that was downloaded from the Internet and viewed through the image viewer. Doppler and

NEXRAD radars provided highly accurate monitoring of the progress of the thunderstorms. In

addition, flood potential was also monitored via the Intemet with FloodCast predictions and

NWS warnings and watches. Also, TVA Lake Information (stream flow rates and heights) was

retrieved to access the flood potential on the Tennessee River.

When it was determined that the Tennessee was exceeding its banks, data providers were

contacted to determine if there was a satellite image that could be used to ascertain the extent of

the flood. It was determined that there were none, but a local TV helicopter crew took some still

photography of the river. This was simulated through the use of aerial photography previously

taken by SENTAR personnel. These images were taken at oblique angles to the ground which

increases the difficulty of georectification. The images were developed and taken to the DPC

where they were scanned, imported into Map-X, and georectified. Next, the images were

| ") 97-0269

Figure 9. In-land wind output for scenario.

Figure 10: Sample of the products produced for Scenario 1.

1 "_ 97-0269

processed into BIL formatted files for the three color bands and a single-band simulated true

color image. Due to the smaller areas of the photographs (generally under 5-10 Mbytes), there

was no problem transferring then via the Internet.

With the new images imported into ArcView, the capabilities of the ArcView tool were again

used to generate outputs. In this case, the original view of the river was represented using the

water classification vector from the pre-flooded Landsat image and the new view of the river

was represented by the aerial photographs as raster images. Water classification vectors were

not generated from the aerial photographs since the oblique angle to the ground was an

unknown, as were the characteristics of the atmosphere. These factors were beyond our

capabilities to make the necessary corrections to account for the color variation across the imagefor a common ground feature.

The FEMA floodplain data was again used to show where the 100 and 500 flood zones are in the

area of concern. Figure 11 shows a sample of the nonflooded Landsat TM data with selected

TIGER data for the affected area. Figure 12 shows the same area with one of the aerial photosadded. Finally, in Figure 13 the water classification vector file for the nonflooded situation is

displayed over one of the aerial photos.

Figure 11. Landsat TM band 7 and TIGER road, hydrology, and railroad data.

|A 97-0269

Figure 12. Landsat TM and TIGER data with aerial photograph.

Figure 13. Figure 12 image with Landsat derived water classification vectors.

| q 97-0269

Other Activities

Home Page. During this milestone period, some further work on the redesign of the projecthome page that was started last milestone period was continued.

Principal Investigator Conference. The Principal Investigator (PI) represented the project at the

NASA/FEMA Conference on GIS and Applications of Remote Sensing to Disaster Management

in Washington, DC, January 14-16, 1997. Appendix C presents the slides that were used at thatconference.

CivilAir Patrol. As a result of contacts made at the September 1996 PI conference, SENTAR

established contact with the Director of Communications of the Civil Air Patrol regarding

integration of their system for near real-time aerial photograph capturing and downloading with

our system for georectification of those photos and integration with the GIS. We met with

representatives of the Alabama wing of the CAP in an attempt to work out a plan for testing this

as part of the prototype exercises/demonstrations, but the aligning of schedules and the

availability of the necessary equipment and personnel did not come together to make thishappen.

Problems Encountered

The following were significant problems encountered during the course of the project that had

impact on the overall project results or outputs.

Data Acquisition. There was confusion at the start of the project about whether project funds or

NASA would pay for the remote sensing data used on the project. The original budgeting for the

project assumed that NASA would supply the necessary data, as the data were coming from

NASA archives. When it was determined that project funds would be required, there were

further delays in determining whether the project could purchase data as a NASA affiliated user.

The result was that actual Landsat TM data were not acquired until well into the second year ofthe project.

Internet Connectivity. Alabama EMA was unable to establish dedicated Internet access through

the Alabama Supercomputer Network operated by the Alabama Supercomputer Authority for the

exercises/demonstrations. Internet access was provided through the MSFC GHCC using their

800 dial-up service. Current plans have delayed the start of Internet connectivity through

Alabama Supercomputer Authority until early 1997.

Clarification of Collaborative Agreements. Some of the details of the required support from

USASSDC were not explicitly defined in the articles of collaboration. This involved equipment

or personnel that were available at USASSDC where SENTAR would operate from and, thus,

were not always identified in the articles of collaboration. When later USASSDC needed to

move the SENTAR office, it was not clear what we needed and we were set up in an area that

did not fully meet our needs. The result was a period of time where there were

misunderstandings as to what was needed to support the project and what could be provided.

User Identification. The final planning and execution of the exercises/demonstrations were

hampered by the fact that Alabama EMA is currently in the process of deciding who should use

the GIS related software tools and how they should be integrated into their operations. It will

take some time before this gets resolved, and until it does it limits the potential use of the tools

|_ 97-0269

and data provided as part of this prototype system. It will probably take approximately one year

of close consultation with the appropriate staff and working actual disasters with them to fully

address how best to fit these tools and data into their operations.

Milestones. The project was created as a series of sequential milestones that were put in place

to ensure that there was regular progress across the project. At times during the project,

particular milestones proved difficult to meet at the preplanned times. Where the completion of

these particular milestones at those specific times was not hampering the progress of the project,

it would have been useful to reschedule the milestones rather than spend excessive project

manpower resources completing a milestone that did not need to be completed at that time.

Requests for Assistance from NASA

No technical assistance requested from NASA GSFC was requested during this milestone period.

FUTURE ACTIVITIES

The future activities for SENTAR related to the NASA CAN are:

1. Solicit Follow-on Funding from Alabama EMA. SENTAR is pursuing Alabama EMA to

fund a continuation of the effort. The primary goal is to complete the effort to work the GIS

and remote sensing products into their specific operational procedures.

2. Take the Prototype System Concept to Other State EMAs. SENTAR is refining what are

the specific tools and data that are required to adapt this prototype system to other state EMAoffices.

, Explore Off-Shoot Efforts Using Remote Sensing Data. SENTAR is pursuing off-shoots

from this project that involve the use of remote sensing imagery. These include other

government research solicitations to extend the specific remote sensing data usage, potential

commercialization with a company producing weather radar with storm tracking software,

and the start of a new corporation from SENTAR to pursue the development of a new type of

Internet-based mapping service that incorporates remote sensing data.

1"7 97-0z69

APPENDIX A

PROJECT SUMMARY

_Sentar Project Summary

Flood Management Enhancement

Using Remotely Sensed Data

Project Summary

Prepared by

Gregory J. Romanowski

Sentar4910 Corporate Drive, Suite C

Huntsville, AL 35805

(205) 430-0860

May1996

Cooperative Agreement No. NCC5-111

Prepared for

NASANational Aeronautics and Space Administration

Goddard Space Flight Center, Code 912Greenbelt Road

Greenbelt, MD 20771

THE VIEWS, OPINIONS, AND/OR FINDINGS CONTAINED IN THIS REPORT ARE THOSE OF THE

AUTHOR(S) AND SHOULD NOT BE CONSTRUED AS AN OFFICIAL GOVERNMENT POSITION, POLICY,OR DECISION UNLESS SO DESIGNATED BY OTHER OFFICIAL DOCUMENTATION.

Sentar

TABLE OF CONTENTS

Project Summary

Section Pa_agf.

INTRODUCTION .......................................................................................................... 1

Background ............................................................................................................ 1

Purpose .................................................................................................................. 1

PROJECT SUMMARY .................................................................................................. 1

Project Goals .......................................................................................................... 1

Project Status ......................................................................................................... 2

REMOTE SENSING DATA ANALYSIS AND PROCESSING ..................................... 4

DAAC Data Research and Processing ..................................................................... 4

Georeferencing Oblique Aerial Photography ........................................................... 9

PROTOTYPE SYSTEM ............................................................................................... 14

Overview .............................................................................................................. 14

Software Modules ................................................................................................ 15

EMO Tool ............................................................................................................ 16

Vector/Raster Data Formatter .............................................................................. 19

Data Retrieval Scripts ........................................................................................... 19

GIS Data Layer Development ............................................................................... 19

FUTURE ACTIVITIES ................................................................................................ 26

Sentar

FLOOD MANAGEMENT ENHANCEMENT

USING REMOTELY SENSED DATA

PROJECT SUMMARY

Project Summary

_TRODUCTION

Background. SENTAR, Inc., entered into a cooperative agreement with NASA Goddard Space

Flight Center (GSFC) in December 1994. The intent of the NASA Cooperative Agreement was

to stimulate broad public use, via the Internet, of the very large remote sensing databases

maintained by NASA and other agencies, thus stimulating U.S. economic growth, improving the

quality of life, and contributing to the implementation of a National Information Infrastructure.

SENTAR headed a team of collaborating organizations in meeting the goals of this project.

SENTAR's teammates were the NASA Marshall Space Flight Center (MSFC) Global Hydrology

and Climate Center (GHCC), the U.S. Army Space and Strategic Defense Command, and the

Alabama Emergency Management Agency (EMA).

For this cooperative agreement, SENTAR and its teammates accessed remotely sensed data in the

Distributed Active Archive Centers, and other available sources, for use in enhancing the present

capabilities for flood disaster management by the Alabama EMA. The project developed a

prototype software system for addressing prediction, warning, and damage assessment for floods,

though it currently focuses on assessment. The objectives of the prototype system were to

demonstrate the added value of remote sensing data for emergency management operations

during floods and the ability of the Internet to provide the primary communications medium for

the system. To help achieve these objectives, SENTAR developed an integrated interface for the

emergency operations staff to simplify acquiring and manipulating source data and data products

for use in generating emergency management data products. The prototype system establishes a

systems infrastructure that is designed for expansion to include future flood-related data and

models or to other disasters with their associated remote sensing data requirements anddistributed data sources.

Purpose. The purpose of this report is to: 1) provide a summary of the accomplishments of the

project and 2) provide some thoughts the future of this work. This information is intended to

provide project insight to NASA and others in applying NASA remote sensing data to disaster

management.

PROJECT SUMMARY

Project Goals. The project goals were to develop a GIS-based, prototype system for enhanced

emergency management of floods that: 1) utilizes Remote Sensing Data (RSD), the Internet, and

existing software products and 2) is readily expandable into an all-hazards tool for use throughout

Alabama or anywhere the necessary source data is available.

The specific objectives were to:

] 96-0108-J

Sentar Project Summary

1. Identify RSD, particularly from NASA's DAACs, that have the potential to enhance flood

management

2. Develop and demonstrate a prototype system for utilizing RSD for flood management byAlabama EMA staff

3. Demonstrate the value added of RSD for emergency management operations

4. Demonstrate the feasibility of using the Intemet as the communication network from theDAACs to end user.

The system concept developed uses remote sensing data sets from the NASA DAACs, or other

data providers, that are integrated into a commercial Geographical Information System (GIS).

When used with the geographical, infrastructure, and demographic data in the GIS, the remote

sensing data provide the Alabama EMA with additional capabilities for addressing flood relateddisasters.

The second element of the system concept is the use of the Internet as a communication path for

transferring remote sensing data and data products from the data providers to the processing

center, and then to the Alabama EMA and its staff in the field. The project showed that much ofthe communications can be achieved via the Internet.

Project Status. At the conclusion of the NASA sponsored phase of this research project, a

prototype system has been developed consisting of four major geographically-separated

components. The major system components are a Data Processing Center (DPC) (located in

Huntsville, AL), the Alabama EMA Emergency Operations Center (EOC) (in Clanton, AL), the

remote sensing data providers (various locations in the world), and the remote or county users (at

various locations within the state of Alabama). For the prototype system, the DPC is split

between two facilities: the NASA GHCC for remote sensing data processing and the USASSDC

for the Geographic Information System (GIS) and ARC/INFO conversion processing. The

specific elements that have been installed and demonstrated are the:

• DPC for remote sensing processing using the commercial software package MapX, a

SENTAR produced data reformatted for importing the files into ArcView, and the

commercial package ARC/INTO

• The Emergency Management Operations (EMO) Tool developed by SENTAR to provide a

single user interface for the COTS applications to be used by the emergency mangers on theirPCs

• ArcView 3.0 for data viewing, manipulation, and spatial analyses by the emergency managers

• Sample data files for the EMO Tool

• TIGER GIS vector data consisting 16 data coverages for 17 northern and 13 southeastern

Alabama counties (see Figure 1 and TABLE 1), plus Redstone Arsenal boundary, all providedon CD

FEMA Q3 Flood data for 19 counties in Alabama (see TABLE 2), provided on CD

One state spatial data coverage for power plants

2 96-0108.J

Sentar Project Summary

Landsat TM imagery for the areas covered by the TIGER GIS data - these images are

provided as multiband images, a three-band composite, and a water classification in vector and

raster formats, provided on a CD for each of the Landsat source images (Figure 2)

Sample USGS National High Altitude Aerial Photography in color and black and white

images for selected high flood areas.

TABLE 1

TIGER DATA COVERAGES

Census block groups

Roads and state highways

Interstate highways

Streams and rivers

Lakes

Airports and airfields

Railroads

State boundaries

Counties boundaries

Urban area boundaries

School districts

104 th Consressional districts

Voting districts

Apartment complexes

Trailer parks

Shopping centers

Figure 1. TIGER data for Alabama counties in the prototype system

3 96-0108-J

Sentar Project Summary

TABLE 2

ALABAMA COUNTIES WITH FEMA Q3 FLOOD DATA

Baldwin Covington Houston Morgan

Barbour Dale Jefferson Randolph

Calhoun Dallas Madison Russell

Coffee Geneva Mobile Tuscaloosa

Conecuh Henry Montsomery

The components of the prototype system that shall be implemented post the NASA sponsored

research project are:

• Loading and configuring the Internet retrieval scripts on the AEMA server once they have

dedicated Internet access

• Integration of the GIS spatial analysis products into the EOC's standard operating procedures

• Expansion of the GIS database with the additional State Agency spatial data

• Consolidation of the DPC into a single facilities.

There was a demonstration of the system capabilities and the databases that have been provided to

EOC staff. This was supplemented with a training session on how the use the data to produce the

data views and spatial analyses needed during and aider a disaster.

From here, the system will be supplemented with additional satellite or aerial photography as it

becomes available or is specifically acquired and additional state spatial data will be added as it

becomes available until all the desired data coverages are included.

REMOTE SENSING DATA ANALYSIS AND PROCESSING

DAA_C Data Research and Processing. SENTAR initially investigated the use of several data

types from NASA's archives. Specifically these include the Advanced Very High Resolution

Radiometer (AVHRR), Special Sensor Microwave/Imager (SSM/I), Geostationary Operational

Environmental Satellite (GOES), and Landsat Thematic Mapper (TM)and Multispectral Scanner

System (MSS). The results of that investigation was that only Landsat TM was desirable for use

in the GIS for water delineation visually and through classification analysis. GOES was

determined to be useful in the product oriented version on the Worldwide Web. Landsat MSS

could be useful if no other data were available, but generally the resolution was too coarse for

emergency management operations.

SENTAR performed a review of the current research in the application of remote sensing data for

predicting, assessing, or monitoring floods. The objective was to determine if there were other

uses of or types of remote sensing data for floods that we had not considered. The relevant

journals, periodicals, conference proceedings, and NASA technical archives were searched for

4 96-0108-J

Sentar Project Summary

papers and reports. Listed below are some of the more significant papers and reports that we

reviewed. Generally, there were no surprises that were uncovered in terms of sensors that could

be used. Some interesting work has been done on using AVHRR data for some special flood

events, but the conditions where the AVHRR data would be useful was atypical for Alabama.

Thus, the satellite data to be used was limited to Landsat TM for incorporation into the GIS for

integration with the GIS data and GOES data for use as weather information in the image formatavailable on the Worldwide Web.

SENTAR processed Landsat TM imagery as the primary imagery backdrop for the prototype

system and produced the data on CDs. The satellite data cover all of northern Alabama and the

southeast portion of the state (see Figure 3). The data were reformatted for importing into the

Map-X remote sensing software were the individual bands were combined into single file and then

converted into a BIL formatted file for exporting to ArcView. The composite band file is a large

file (-300 Mbytes) that is cumbersome for the emergency managers to work since all six bands

are needed. Therefore, SENTAR also produced simulated true color single-band files. For these

files, three of the Landsat TM bands were combined into a color composite image and then

converted into a single-band file. A Map-X algorithm was applied to enhance the colors of the

composite file. This algorithm generate a simulated true color image by analyzing the range of

pixels values used in and then spreads out the color values across the 256 bits of each pixel of the

composite file. Figure 4 shows an example of the effect of the simulated true color image

processing. The resulting file was also converted into a BIL format for exporting to ArcView.

Figure 3. Area covered by the processed Landsat TM data.

5 964) 1084

Sentar Project Summary

Finally, water classification raster files were produced from the Landsat image using the tools

within Map-X. A median filter was applied to the resulting image file to eliminate solitary pixels

identified as water. Although these may, in fact, be water, the desired output is for significant

bodies of water and not every small pond. The boundaries for the water classified areas were then

converted into a vector representation from the raster image. The vector file was converted into

CAD format (.dxf) for exporting to ARC/INFO. Once in ARC/1NFO, the vector file was again

converted, this time to the ARC/INFO .e00 file format and ArcView .shp file formats. It is these

last formats that were exported to Alabama EMA along with the original raster band images and

the simulated true color image.

These various image and vector files were grouped by the Landsat TM image from which theywere derived and were burned onto a writeable CD. A readme file was also added to describe the

contents of the CD and a Process.txt file was added to describe the processing performed on the

original data to produce the data sets on the CD. In addition, an image file was added to show

the area extent of the four Landsat TM images and which specific one is on the CD (see Figure

5). This format for the files on CD represented a prototype of what production CDs consist of.

As other satellite images become available for a disaster they will be similarly processed and

packaged on a CD. This same procedure would apply to SPOT images and those of the new US

commercial satellites that will becoming operational in late 1997 or 1998.

6 96-0108-J

Landsat TM image bands 4 (red), 3 (green), and 2 (blue)

_,_ _

Map-X Simulated True Color Image of above Landsat TM image

Figure 4. Example of the Map-X simulated true color image processing.

7 ,,_,-ol o_ J

Figure 5. Image on CDs to visualize the areas of the Landsat TM images.

8 _o !)]08-J

{ Sen tar Summary

SENTAR also processed some selected U.S. Geological Survey (USGS) National Aerial

Photography Program (NAPP) and National High-Altitude Photography (NHAP) program

photographs to use as base map images for narrowed waterways that regularly flood to provide

higher resolution than the Landsat TM images. These are hardcopy images that were scanned

into a digital format for processing. Some black-and-white and color images were processed into

BIL files for exporting to ArcView. The photographs acquired cover areas of southern Alabama

that have experienced major floods in recent years. These images cover ground areas of 5x5

miles or 8x8 miles. Figure 6 shows a sample of two of these images with some TIGER GIS data.

Figure 6. Example of the USGS NAPP photographs with some TIGER GIS data

Georeferencing Oblique Aerial Photography. Because of the limitations on the availability of

satellite images when they are needed, SENTAR researched the use of non-professional aerial

photography. The limitations of the available satellite imagery is most notably the availability of

the satellite over the site of interest at the time of interest, lack of visibility due to cloud cover, the

processing and delivery time from the satellite data producers, the resolution of the imagery, and

the cost of the imagery. The use of aerial photography has the potential of eliminating most of

these limitations, although professional quality aerial photography is equally expensive as the

satellite imagery. An alternative is the use of non-professional photography, for example would

be photography taken by the Air National Guard which already supports the Alabama Emergency

Management Agency, or even amateur photography taken by anyone with access to an aircraft

and a camera. A critical question with this approach was the ability to georeference photographs

taken at oblique depression angles for integration into the GIS.

9 96-0108J

Se_ gt_r Project Summary

To address this question, SENTAR performed a short parametric analysis on a series of aerial

photographs taken aboard a small aircraft using a standard 35 mm camera. The analysis looked at

aerial photography at altitudes of 1000 to 9,000 feet above the local terrain, at a series of

depression angles, and at four geographical positions relative to a fixed ground location. These

photographs were then scanned into a digital format, enhanced for visibility, and then

georeferenced. As the depression angle increased, particularly for the higher aircraft altitudes, the

distortion of the georeferenced image increased. While this was expected, the more interesting

conclusion was that there was sufficient accuracy in the georeferencing to meet the needs of

emergency management operations. Figure 7 show the original and georeferenced image

photographs at a depression angle of about 40 °. The before image is as seen in the original

photograph and the after is the processed image with overlaid road and hydrology vector files..

The distortion in Figure 7 is clearly evident as the distance from the camera to the ground

increases and the ground area per pixel increases. In addition, the accuracy of the georeferencing

in these portions of the image is reduced. The overlaid road and hydrology vector files are clearly

less accurate in the lower portion of the georectified image. Note that the matching of the vector

files and the image is not exact, but is clearly close enough to perform a graphical analysis of the

effect of flooding, as depicted in the image, to the various vector data layers.

One limitation of the aerial photography versus the satellite images is the reduced areas of

coverage. This was particularly true for the test photographs taken by SENTAR. We were using

a light aircraft with a ceiling of 10,000 feet above sea level (or 9000 feet above the local terrain

for the area photographed). To compensate for this and also to remove some of the heavily

distorted areas of an image, multiple georeferenced images can be combined in a mosaic, as

shown in Figure 8. Here three images of the same area have been combined to produce a

composite that provides a wider area of coverage. The three images used for this composite

image were taken at altitudes of 3 000 and 9000 feet above the local terrain; aircraft bank angle of

0 °, 20 °, and 45°; aircraft headings of south, north, and east; and a camera depression angle of

approximately 40 °. Figure 9 presents a slide that was developed for briefings that summarizes the

methodology used to create the image of Figure 8. A similar approach can be taken to create a

mosaic image that covers a length of a fiver or an extended area.

The intent is to export these images to the EOC where they will be used within the GIS with the

feature data layers. They can also be combined with other images, such as hardcopy maps to

produce static maps for distribution to those without the GIS software. Figure 10 provides an

example of this for one of the aerial photographs with vector data for the roads and hydrology

and the USGC hardcopy map for those areas beyond the limits of the photographic image.

10 96-0108-J

Sentar P_oj_c_S_a_

Figure 7. Original and georeferenced aerial image of a portion of the Flint River (taken at

an altitude of 10,000 feet above sea level and at an aircraft bank angle of 0 °)

I I 96-0108-J

Ser_ tar t,roject Summaw

Figure 8. Mosaic image of three aerial photographs taken at different aircraft altitudes,

headings, and bank angles.

2 96-0108-I

Sen tar Project Summary

!Merging Oblique Photographsf_Om a hand.held,, 35_,_ cBmera

in a |_ght a_rc_aft

: _-_ _

._ r .¢- : ".

Figure 9. Slide summarizing the process of mosaicing three images to create a composite of

the aerial photograph images.

A side benefit of the aerial photographs it that provides current information on some of the

feature data of the GIS data layers as well as providing information on the effect of the disaster.

Specifically, in the test images along the Flint River, we were able to see how roads have been

changed or added. Similarly, changes in streams were noted, primarily due to manmade changes

for road construction or agricultural reasons. This side benefit is significant because it can

provide emergency management staff with more accurate representations of the area than in the

feature data layer and because it can provide those familiar with the area with greater confidence

in the GIS-based system. When the GIS data layers do not accurately reflect the spatial feature as

the user knows it, the user loses confidence in the entire data layer and eventually the systemitself.

13 96-0108-J

Sentar p,-oj tSummary

i

Figure 10. Example of a static map composed of an aerial photograph, USGS hardcopy

map image, and vector feature data for roads and hydrology.

The use of satellite and professional imagery will always be a primary source of remote sensing

data where available, however this effort has also shown that the use of non-professional, oblique-

angle aerial photography also can be of value to emergency managers.

PROTOTYPE SYSTEM

Overview. The prototype system developed by SENTAR and its collaborators for integrating

remote sensing data into a GIS for emergency management operations is illustrated in the Figure

1 1. The central components of this system are the Emergency Operations Center (EOC) and the

Data Processing Center. These two centers are supported by various data sources and the EOC is

itself a data provider to the County Emergency Management offices, remote filed personnel,

FEMA, and even the general public. The Data Processing Center produced a baseline set of GIS

data coverages from US Government agencies and state sources and a set of Landsat TM images

from the EROS Data Center (one of the EOSDIS DAACs) for the regions of the state covered by

the prototype system. These data sets will be supplemented with other remote sensing data

products as disasters are encountered. Additional data sets for actual floods may come again

from the EOSDIS or from other satellite data providers or from aerial photography. These data

14 96-0108-J

Sentar Project Summary

that are to be georectified will be processed and reformatted at the Data Processing Center and

then sent to the EOC for usage by the emergency managers. Remote sensing data not requiring

georectification or reformat processing will go directly to the EOC for usage.

At the EOC, the database of the baseline GIS and remote sensing data is supplemented with

additional GIS data coverages generated from spatial data provided by other state agencies. The

emergency managers use the baseline and current data sets within their GIS software and other

viewing tools to generate image analyses of the potential flood area, the actual flood area, or

flood damage assessment. By combining the GIS and remote sensing data, they are able to obtain

information not possible with only one data type. The views of the combined data sets they create

can be captured and sent to County Emergency Management offices, remote field_personnel, and

FEMA, as appropriate. In the future, it is planned to also provided selective information directly

to the public via a Web site and via the state television broadcasters.

The primary tool the EOC staff will use is ArcView, a commercial GIS software package. In

addition, they will access various Internet sources for Web-based information, FTP and Gopher

sites, and contact with other emergency management professionals. SENTAR has developed a

graphical user interface to facilitate the accessing of the various applications for ArcView, the

Internet retrieved data, and other PC-based tools that they currently use through a single

interface. This interface application is called the Emergency Management Operations (EMO)Tool.

Software Modules. The prototype system consists of a set of commercial-off-the-shelf (COTS)

applications integrated under a common user interface, called the Emergency Management

Operations Tool. The system makes extensive use of existing software products at the Alabama

EMA to provide system functionality wherever possible. Where these software packages did not

meet the functional needs of the prototype system, SENTAR developed software applications and

utilities. Three software components were developed for this project, they are:

1. Emergency Management Operations (EMO) Tool

2. Vector/Raster Data Formatter

3. Data Retrieval Scripts.

The EMO Tool resides on the PCs of the EOC staff, the vector/raster data formatter operates on

PCs at the Data Processing Center, and the data retrieval scripts operate on the Sun server at the

EOC; refer to Figure 11.

1 5 96-0108-J

Sentar Project Summary

.__. F_D

.// _ I _''ur_z I

TV& N _ir3,S_ _ PrD_ " Ne_d _ O_ezD_la _._o_s_-,g . -.__1 Feal_e [i_r_i_ Cer_ _/_at_r_l_s

I ¢T__ I t___-, A i /'

I s_oa_ IA_" I

_B3C ... IP_ ." I

/

Ra D6 I_÷6i_$:_n. _._r_ 9, _', _r_e_n'e_ _iDrn'_licm _'oin',_zje_

OB:e

Le_r,di

IVteOia

Ce_l_sC<_runic:a$:_s

L

Fa=

IInle_rel

Figure 11. Prototype system concept.

Two of the three software components were installed on the appropriate platforms. These consist

of the EMO Tool and the Vector/Raster Data Formatter. The Data Retrieval Scripts were not

actually installed on the Alabama EMA server for testing as they did not have direct Internet

access for the demonstrations. Instead, these were run form SENTAR's ISPs computers.

EMO Tool. The EMO Tool provides the EOC staff with a single user interface for the COTS

applications integrated into the prototype system for access by the operations staff from their

PC's, thus allowing separate COTS applications to be integrated into one package. The interface

to the EMO Tool easily allows the EOC staffto acquire and view data downloaded offthe

l 6 96-0108-J

Sentar Project Summary

Internet (such as GOES, and radar images along with NWS forecasts and warning bulletins). The

tool also executes and spawns external applications at the touch of a button, most notably the

ArcView GIS software, without having to locate icons within program groups or search in

directory paths.

The prototype tool was developed in Visual Basic 4.0 as a 16-bit application and runs under the

Window 3.1 and Windows 95 operating systems. The tool has been composed with pre-

developed drop and drag components, making future enhancements easy to add on. Figure 12

shows the EMO Tool display with samples of retrieved image and text files.

The tool has been broken down into 4 modules with the first being the graphical user interface

(GUI) (see Figure 12). The GUI is composed of a variety of push buttons and pull down menus.

These buttons and menus give choices to the staff as to what area they might be interested in

within the state of Alabama for viewing radar or weather bulletins from that location.

The second module is an internal viewer for the tool. Although the tool supports an external web

and resource browser such as Netscape or Mosaic, development called for the operation of a

Figure 12. Prototype of Emergency Management Operations Tool

1 7 96-0108-J

Sentar Project Smnmary

hands-off internal viewer. The internal viewer supports image formats in the form of GIF, JPEG,

and TIFF and ASCII text file formats from both PC and UNIX computers. The tool allows the

user to bring in multiple images/ASCII text files each having its own scrollable window. This

multi-image/text ability allows the user to view several files at the same time.

The third module is the internal source code for the image and text file retrievals and routine pop,

pull and refresh code for the windows functions. The source data file retrievals are performed

outside the tool application on the SUN server, utilizing the data retrieval scripts which are

described later. The EMO Tool instead takes images and text files from designated storage

locations on the server and load them into the internal viewer. The reason for extracting these

files from the server rather than directly from the remote computers, is to conserve the bandwidth

of the EOC's Internet connection and to insure that common versions of the files are used by all

emergency managers.

The fourth and final module is the help utility. As with many commercial packages, help files

come with separate help documents separate from the application itself. Though the use of the

file menu system in the EMO Tool, internal on-line help information is supported. These files

provide instructions for each button and menu and provide the user with pictures of how to

operate the tool. The software for the help file functionality is provided by the VB HelpWriter

utility. Figure 13 shows a sample of the help utility display for the icon buttons.

lCOIqS

_ . :: _ii_::.. _;q_:i_.;!_::,-i

The Space Science and Engineering Center (SSEC) is a global leader in the analysis and distribution

of geostationary satellite data. The SSEC is also a part of the University of Wisconsin-Madison'sGraduate School. SSEC is concerned mainly with scientific research and technology developmentthat enhances understanding of the atmosphere of Earth and also looking down to gain informationand insights into weather, climate, and other aspects of our global environment.

The SSEC provides real-time data for visible and infrared images from the (GOES) satellite. Thefollowing GOES images are:

North America GOES V'mible (updated every 30 minutes)North America GOES Infrared (updated every 30 minutes)

Doppler Radar detects the presence and location of an object by bouncing an electromagnetic signal

off of it and measuring the time it takes for the signal to return. This measurement is used todetermine the distance and direction of the object from the radar. In the case of radar meteorology,the "objects" being measured are the particles of water, ice or dust in the atmosphere. Doppler radarstake additional advantage of the fact that radar signals reflected from a moving object undergo a

change in frequency related to the speed of the object traveling to or away from the radar antenna.

Figure 13. Sample display from the EMO Tool help utility.

18 96-0108-J

Sentar Project Summary

Vector/Raster Data Formatter. SENTAR is using of the Map-X remote sensing software to

perform the image processing functions ofgeoreferencing and feature classification. Map-X's

ability to perform 3rd-order georectification provides the SENTAR team with the ability to work

with imagery from high-end satellite data down to low quality 35m aerial photography on a

standard PC platform. The limitation to using Map-X was the inability to export processed

images into the ARC/INFO and ArcView packages while retaining the georeferencing information

because of proprietary file formats. With help from the Map-X developer, Delta Data Systems, in

terms of providing the output file format information, SENTAR was able to develop thevector/raster formatter.

The vector/raster data formatter converts raster imagery from Map-X format to the common

Band Interleaved by Line (BIL) raster file format for both single and multi-band images. The

formatter creates various description files associated with the BIL file: a header file that describes

the layout of the image pixel data, a color file that describes the image color map, a statistics file

that describes image statistics for each band of the image, and two other text files that break down

the internal components of the image file and also describe the format of the first channel.

The formatter was developed using the Visual C++ language. It takes full advantage of'new

input/output stream functions in C++ due to the size of'the complex imagery. The utility itself

goes by the name AIF2BIL because of its creation ofa BIL file from Map-X's .AIF file.

Data Retrieval Scripts. SENTAR's design concept makes extensive reuse of existing software

products to provide system components wherever possible. In the case of data retrieval scripts it

was in the best interest of SENTAR to reuse pre-developed scripts and utilities from the

University of Texas and RSPAC.

The data retrieval script developed by SENTAR is a UNIX-based script which will reside on the

SUN server at the EOC. The script is composed of function calls to a small Perl utility called

url_get. Url_get was written to retrieve documents specified by their Uniform Resource Locator

(URL) on the World Wide Web. Through these calls, the script download radar and satellite

GIF/JPG images, along with gopher NWS forecast ASCII text pages. Other portions of the script

use the standard File Transfer Protocol (FTP) to download files from those sites that support

direct anonymous FTP services. All of the ASCII text files and images will be stored on the EOC

server. These files will then be retrieved from the server by the EOC staff via the EMO Tool.

The retrieval scripts primary purpose is to help in the automatic retrieval of specified files, while

conserving bandwidth used by EOC staff during emergency management operations. The script

will be placed in the EOC server's cron tab giving only the system administrator the proper

permissions on how often it should be executed and which URLs are to be downloaded. This

module is complete except for inclusion of the final set of URLs and final testing on the EOC

server.

GIS Data Layer Development. The SENTAR staff worked closely with the Alabama EMA

staff to determine the spatial data necessary to support emergency management operations for all

disasters. TABLE 3 identifies these data categories, or layers, and groups them under 13

thematic areas. The data layers and the associated data attributes are listed in TABLE 4. The

data layers and attributes per data layer were agreed upon by the appropriate staff at Alabama

EMA. During the final review of the data layers, the Alabama EMA staff identified potential

source departments and agencies for the data layers that consist of state controlled or generated

19 96-0108-J

Sentar Project Summary

data. SENTAR supported Alabama in attempting to acquire the data from the appropriate state

agencies. Telephone contacts were made to verify that the state agency have the needed data and

to begin the process of characterizing the data currency, format, and storage media. Most of the

data are available, but the quality and currency varies significantly. In addition, most of the data

are not in electronic format and almost none of it is in a spatial database. Several agencies stated

that they are now in the process of designing a spatial database for the data, however, none was

available during the duration of this project. Thus, what the data was aquired was manually

incorporated into an ArcView data coverage.

TABLE 3

THEMATIC GROUPS OF DATA LAYERS

THEME

Hydrology

Utilities

DATA LAYERS

FEMA Flood Maps

Hydrography

Lake Levels

Streamflows

Wetlands (D)

USACE Flood plains

Snow & Rain Accumul. (D)

Coastal Hydrology

EAS Radio Coverage

Doppler Radar (D)Bum Sites

Sewer Systems

Fresh Water SystemsFuel Facilities

Water/Sewer

Treatment Plants

Transmission Lines (D)

Electric Power grid (D)Solid Waste Sites

Geology Surface Geology (D) Elevation Contours

Boundaries Military Installations Local Government Bound. Facility Bound.

Built-Up Areas State Administrative Bound. Military Leased Lands

County Bound. Federal Administrative Bound. Restricted Areas

International Bound. Census Geography

Socioeconomic Demographics (population) Demographics (over 55) Economic (income) (D)

Geodetic USGS Control Points

Hazards/ Hazardous/Toxic/ Medical Wastes

Environmental Radioactive Wastes Hazardous Materials

Man-Made Features

and StructuresNuclear Plants Bridges Pipelines

Fire Departments Airports Railroads

EOCs Hospitals Power Plants

Roads & Streets Police Departments Gov't Man-Made

Maritime Ports Designated Shelters Structures

Photogrammetric Photos and Imagery

Cultural Resources Historical Standing Sites and Graveyards

Land Use/LandLand Cover/Land Use

Cover

Emergency Planning Evacuation Routes Disabled, Nursing HomesEnhancements

Remote Sensing Water Classification Flood Warnings Surface Temperature

TVA Lake & Streamflow Data Snow Cover

20 96-0108-J

Sentar Project Summary

TABLE 4

PROTOTYPE SYSTEM GIS DATA LAYERS

Priority Data Layer Data Source

[see note] Prima_ Secondary

1. Hydrology

1 Hydrography TIGER USGS DLG

1A FEMA Flood Data FEMA Insurance

Institute

1A USACE Flood Plain USACE

2 Coastal Hydrology

Deferred Wet Lands

SeaLAB of

Dolphin IslandUS Fish &

Wildlife

Deferred Snow and Rain Accumulation. DLG

2. Utilities

1 Fuel Facilities ADEM

USGS

TIGER,Counties &

Cities

1 EAS Radio Coverage State NOAA

1A Fresh Water Systems ADEM Counties &Cities

1A Sewer Systems ADEM Counties &Cities

1A Water/Sewer Treatment Plants ADEM Counties &

Cities

2 Solid Waste Disposal Sites ADEM

ADEMBurn Sites

Counties &

Cities

Counties &

Cities

Deferred Transmission Lines FEMA DLG

Deferred Electrical Power Grids

Deferred

3. Geology

[Doppler Radar

1 Elevation Contours USGS DEM DMA DTED

Deferred Surface Geology State, USGS

Maps

Attributes

No,I Names

2 Feature Name, Census

Feature Class Code (CFCC)

6 Area, Map Parcel, Flood

Way, Hazard Zone, Flood

Elevations, FIPS Code

4 Area, Hazard Zone, Flood

Elevations, FIPSTBD based on source data

6 Type (well, tank farm,

storage), Capacity, Capacity

Units, Contents, State

(solid, liquid, gas), By-

products

2

5

4

Transmitter Name,

Address, POC, Telephone #

Type (other, water lines,

aqueduct), Diameter,

Facility Name

Diameter, Facility Name

Name, Address, Type,

Diameter, Owner

Type (public, private)

Waste Type (inert, tires,construc-tion debris,

concrete & block), Facility

Name, Address

6 Type (public, private),

Waste Type (medical, POL,

other), Facility Name,Address

2 Type, Elevation

21 96-0108-J

Sentar Project Summary

4. Boundaries

County Boundaries

1A

State Admin. Boundaries

TIGER USGS DLG

1 TIGER

1 International Boundaries TIGER USGS DLG

1 Local Boundaries (cities, etc.) TIGER

Federal Boundaries (parks, etc.)

Military Installation

AL Conserva-

tion Department

MilitaryInstallations

TIGER

TIGER

State, MilitaryInstallations

Built-Up Areas

Census GeographyRestricted Areas

USASSDC,

!USGS, Maps

Counties &

Cities

3A Facility Boundaries Gov/MilitaryInstallations

3A Military Leased Land Military NoneInstallation

5. Socioeconomic

2 Demographics (population) TIGER USGS DLG

2 Demographics (over 55) TIGER USGS DLG

Deferred Economics (income)

6. Geodetic

2 USGS Control Points USGS

7. Hazards/Environmental

1A Hazardous/Toxic/Radioactive ADEM Counties, Cities

Waste & MilitaryInstallations

1A Hazardous Materials (also ADEM Counties, Cities

private incinerators, medical & Military

waste) Installations2 Medical Waste ADEM

8. Man-Made Features and Structures

1 Roads and Streets iTIGER USGS DLG

1 Airports DMA DAFAF USGS DLGRailroads

Nuclear Plants

TIGER

State

1 EOCs AEMA

USGS DLG

EIS, FEMA

2 State, County (FIPS)

3 State, Type, CFCC

2 Type, ID Code

3 Local Name, County, StateTBD based on source data

2 Type, ID Code

1 Name

TBD

4 Area Name, Managing

Agency, Type

2 Name, Type

2 Type, ID Code

5 State & County Names,

Block & Group, Number,]Housing Units

5 State & County Names,

Block & Group, HousingUnits

Income and House Values

1 ID Number

EPA Number, Facility

Name, Address

EPA Number, Facility

Name, Address

CFCC, Feature Name,

Feature Type, Alt. FeatureNames

Feature Name, No-Fly Zone

CFCC, Feature Name,

Feature Type

Site Name, Category, Type

(technology), Buffer Dist.,

Operations Phone #, StreetAddress

EOC Name, Address,Phone #

22 96-01084

Sen tar Project Summary

1 ! Hospitals AL Dept. of FEMA, TIGERPublic Health

1 Designated Shelters AEMA

1A Bridges AL DOT

2 Pipelines County, City, Corps of

State, Military EngineersInstallations

2 Power Plants State, TVA, Counties &

APC, Co-Ops, Cities

MilitaryInstallations

2 Government Man-Made

StructuresState, MilitaryInstallations

3 Maritime Ports SWSA

3 Police Departments State, Counties& Cities

3 Fire Departments AL Fire College

Counties &

Cities

Counties &

Cities

9. Photogrammetric

1 Photos and Imagery SPOTEDC, AircraftPhotos

10. Cultural Resources

3 Historic Standing Sites and

Graveyards

Historical

Commission

National

Historic

Registry

11. Land Use/Land Cover

3 Land Use/Land Cover USGS DLG Counties &

Cities

12. Emergency Planning Enhancements

1 Evacuation Routes AEMA County EMAs

1 Disables, Nursing Homes AL Commission

Jfor the Aging

13. Remote Sensing

1 Land/Water Classification EDC, Aircraft SPOTPhotos

4

4

3

4

Name, Operations Phone #,

Address, Number of Beds

Shelter Name, Hazard

Type, Address, Phone #

Load Capacity, OverheadClearance, Elevation

Owner, Type, Diameter,Contents

Facility Name, Category,

Owner, (classification),

Address, Phone #

Description, Owner, Type

(city, county eng office,

county courthouse, Gov

level, etc.) Address, Phone#

1 Name

3 Dept. Name, OperationsPhone #, Address

Type (public, volunteer,

private, commercial),

Name, Operations Phone #,

Capabilities (bomb,

HAZMAT)

Source, Image type (picture,

satellite, movie, TV),

Registration Lat./Long.,

Resolution, Tile #, ImageDate

Facility Name, Owner,

Phone #, Type (including

maritime sites, prehistoric

archeological sites, Native

American sites)

1 Type

1 Route Type

3 Facility Name, Owner,Phone #, Number of Beds,Address

J 5 J Class Type, Location,Source, Projection, Date

23 96-0108-J

Sentar

1 TVA Lake & Stream Flow Data TVA

1 Flood Warnings NWS Floodwatch

2 NOHRSCSurface TemperatureSnow Cover NOHRSC

Note:1

2

3

A

Deferred

Dropped

Priority definitions

Top priority, must be included in initial version

Desirable for initial version if data availability and resources permit

Lowest priority, deferred to a follow-on effortIf available

Lesser priority data layer; deferred to another effortDetermined not to be needed at the installation EOC

Project Summary

5 Sensor Type, Name,

Location, Sensor Value,

Date, Time

4 Source, Type, Start Time/Date, End Time/Date

Temperature

Cover Type

The desired data layers of TABLE 3 were divided into three categories: Those that can be

produced from the U.S. Census Bureau's TIGER data, those that can be obtained form other

federal agencies, and those that will be obtained from state sources. TABLE 5 lists the data

coverages that were produced for the prototype system from the desired data coverages for these

three categories.

SENTAR contracted out for new sets of these data layers using the most current version of

TIGER data. These data layers were supplemented with data layers that were provided by the

U.S. Army from the Integrated Disaster Planning Package (e.g., the Redstone Arsenal boundary).TIGER data for the counties of northern Alabama and southeast Alabama were combined and the

composite vectors were processed to correct for proper edge matching at the county boundaries,

dangling vectors, and open polygons. The completed data coverages were delivered to SENTAR

and a quality check of the data was performed prior to installing the data coverages onto theAlabama EMA server.

The TIGER-derived data layers are essentially the data layers that were provided by the US Army

from the Integrated Disaster Planning Package (IDPP) which used as a starting point for the data

coverages on this project.. SENTAR contracted out for new sets of these data layers using the

most current version of TIGER data. The TIGER data for the counties of northern Alabama and

southeast Alabama were combined and the composite vectors were processed to correct for

proper edge matching at the county boundaries, dangling vectors, and open polygons. The

completed data coverages have been competed and provided to SENTAR.

The next category are the other federal agencies. For these data layers the desired data are either

already on-line or available via electronic media with the except of the Flood Insurance Rate Maps

(FIRM). FEMA developed digital FIRMS for 19 counties in Alabama, which are part of the

Digital Q3 Flood Data. These data were purchased and received from the FEMA Map Center

during the sixth milestone period. All 19 counties will be installed on the Alabama EMA server

during the coming milestone.

24 96-0108-J

Sentar

TABLE 5

DATA LAYERS BY BROAD DATA SOURCES

Project Summary

Data Source Data Layer

TIGER

Other Federal Agencies

State Agencies

Hydrology, boundaries (local, county, state, military,

and international), urban areas, block group, voting

district, school district, demographics (census tract),

interstate highways, roads and streets, railroads,

airports, runways,

Flood Insurance Rate Maps, elevation contours,

surface geology, USGC control points, land use/land

cover, surface temperature, snow cover, flood

warnings

Fuel facilities, EAS radio coverage, fresh water

systems, sewer systems, water/sewer treatment plants

, solid waste disposal sites, burn sites, federal

boundaries, hazardous/toxic/radioactive waste,

hazardous materials, medical waste, nuclear plants,

EOCs, hospitals, designated shelters, bridges,

pipelines, power plants, police and fire departments,

historic standing sites and graveyards, evacuation

routes, nursing homes

For the state agency data, Alabama EMA sent out formal letters requesting the desired data.

Responses from some of the agencies have been received. These were followed up with contact

by SENTAR to discuss the details of the proposed data transfers. Data has been provided by

Alabama Power commission and the construction of the power plant and nuclear power plants has

begun. Additionally, the Alabama Historical Commission is collecting the requested data for the

historical standing sites and will release data to the Alabama EMA when compiled. The Alabama

Department of Environmental Management (ADEM) has agreed to the release of their data,

which will provide data for most of the state data coverages. They are, however, now in the

process of putting all their data into a GIS, so the delivery of the requested data will be delayed

until the new ADEM GIS is built.

25 96-0108-J

SentarProject Summary

FUTURE ACTIVITIES

The primary goal for future activities is to turn this prototype system into an actual operational

system at Alabama EMA, and then at other state or county EMA offices. The specific activities

to be performed are:

1. Work with the EMA operational staff to define specific outputs from the integration of the

remote sensing data and GIS data in ArcView based on their standard operation procedures

for floods or other disasters.

2. Expansion of the data coverages for those data categories that are controlled by the state

agencies.

3. Build an archive of processed satellite and aerial, photography.

4. Refine the satellite image files that are packaged on a CD to meet the specific usage needs of

the EMA staff; this may mean providing single band files or subsetting the Landsat image into

two or four images for easier handling.

5. Pursue the initial efforts to support more predictive tools by combining of soil moisture,

precipitation data, and river or lake levels; with particular attention to the use of the greater

detail digital elevation model data now becoming available.

_6 96-0108-J

APPENDIX B

CD README AND PROCESSING DESCRIPTION FILES

CD Contents Listing in Readme.txt file:

SENTAR CD97-001 April 1997

Landsat TM Image Data

Path 19/Row 38, Acquisition date: April 10,1993

This file should be viewed using a non-proportional font, such as Courier.

This CD contains raster and vector data products derived from Landsat TM

imagery for southeast Alabama. These products are formatted for use in

ArcView 3.0 or compatible software. Files specific to the software which

was used to do the data processing (MapX by Delta Data Systems,

tel. 601-799-1813) are in the MAPXFILE folder.

This is one of a series of 4 CDs. INDEX.GIF shows the location of this

data set and the other satellite imagery in the series with respect to the

state and county boundaries.

The data processing was done by SENTAR, Inc. with the assistance of

the Global Hydrology and Climate Center, which is part of NASA's

Marshall Space Flight Center (MSFC). This work was sponsored by a NASA

Goddard Space Flight Center (GSFC) Cooperative Agreement and is intended

to support the Alabama Emegerncy Management Agency.

For a discussion of the data processing see the file PROCESS.TXT

Points of Contact:

Greg Romanowski and Greg Hodges, SENTAR, Inc.

2 97-0269

4910CorporateDrive,SuiteC, Huntsville,AL 35805

205-430-0860, sentar@iquest.com

Doug Rickman, Global Hydrology and Climate Center

977 Explorer Blvd.

Huntsville, AL 35806

205-922-5889, doug@hotrocks.msfc.nasa.gov

CD TABLE OF CONTENTS

INDEX.GIF TM image (band 5) with state and county boundaries.

PROCESS.TXT Description of data processing used.

README.TXT This file.

ALLBANDS Folder

TM019038.BIL

TM019038.CH1

TM019038. CLR

TM019038.HDR

TM019038. TXT

Six band image of TM bands 1-5 and 7 with raster data

(bil), color table (clr), channel 1 (chl), header (hdr),

and text (txt) files

Projection is UTM and Datum is NAD 1983

ARC FILE Folder

TM019ARC.DBF

TM019ARC. SI-IP

TM019ARC. SHX

TM019PLY.DBF

TM019PLY. SlIP

TM019PLY. SHX

TM019038.E00

Water Classification folder of ARC/INFO vector files

Water classification files for bands 1 and 6 in

ARC/INFO export format (.e00) and

ArcView shape file formats for arcs and

polygons(*arc.* and *ply.*, respecitively)

Projection is UTM and Datum is NAD 1983

IMAGES Folder

STCI1938.TIF

STCI1938.JPG

Simulated True Color Image for bands 7-5-1, R,G,B

in TIFF and JPEG image formats

3 97-0269

MAPXFILE Folder

empty

STCI Folder

STCI1938.BIL

STCI1938.CH1

STCI 1938. CLR

STCI 1938.HDR

STCI1938.TXT

Simulated True Color Image, bands 4, 3, and 2 for red,

green, blue with raster data (bil),color table (clr),

channel 1 (chl), header (hdr), and text (txt) files

WATER CL Folder

SCT 1938.BIL Water classification generated from scattergram from

SCT_1938.CH1 Bands 1 and 6,

SCT_1938.CLR with raster data (bil), color table (clr),

SCT_1938.HDR channel 1 (chl), header (hdr), and text (txt) files

SCT 1938.TXT

4 97-0269

Landsat TM Data Processing Description in Process.txt file:

April 1997

Landsat TM Image Data Processing

SENTAR purchased and processed Landsat TM imagery for use by the Alabama

Emergency Management Agency. The processed Landsat TM imagery is intended

to be used as backdrop images within a vector GIS by showing ground vs.

water during non-disaster (flood) conditions. The imagery has also been

processed using a two-band scattergram to produce for water classification

and vector representations.

The satellite data processed consist of four Landsat TM images. These

cover potions of northern Alabama and the southeast and south central

portions of the state. The raster image processing and conversion from

raster to vector was done using the Map-X remote sensing software, made by

Delta Data Systems, tel 601-799-1813.

1. The source data were reformatted into Map-X format. This consisted of

reformatting the individual satellite bands and combined them into single

file of all six bands. This file was subsequently converted into BIL

format for exporting to ArcView and is contained on this CD in the ALLBAND

folder.

2. Three of the six bands were combined in Map-X to make a color composite.

A Map-X algorithm was applied to generate a single band (8 bit) image that

simulated a true color (24 bit) image. This was also exported into a BIL

format. This file is contained in the STCI folder.

3. A water classification was produced on the Landsat image using a

two-band scattergram within Map-X. The basis for this step is the fact

5 97-0269

that water is darker in the two TM bands used than other land covers. For

some images, this raster classification file was further processed using a

median filter to eliminate single water pixels (presumably ponds and/or

image noise) throughout the image. The final raster file containing the

water classification was converted to a BIL format and is in the WATER CL

folder.

4. The boundary for the water classification was then converted into a

vector representation from the raster image. The vector file is converted

into the CAD format .dxffor exporting to ARC/INFO. The vector files of

the water classification in Map-X and CAD formats are contained in the

MAPXFILE folder.

5. Once in ARC/INFO, the vector file is again converted. This time to the

ARC/INFO.e00 format. It is this last format that can exported to ArcView.

The ARC/INFO and ArcView formatted versions of the vector water

classification are contained in the ARC FOLE folder.

6. Images of the single band, multi-band Landsat TM data and the Map-X

scattergrams are also provided in non-georeferenced TIF and in the JPEG

format. These are in the IMAGE folder.

6 97-0269

APPENDIXC

NASA/FEMA CONFERENCEON GISAND APPLICATIONSOFREMOTESENSING

Marshall SpaceFlight Center

Application of Satellite and Aer_

to Enhance Emergency ManagOperations for Floods

magerye

Presentation to the

Conference on GIS

Sensing to

and Applications of Remote

Disaster Management

January 15, 1997

© SENTAR, Inc. 1997

S, E,N, T,A, R, Inc,Information Technology Solutions for Complex

97-0133

EEIO-L6

,,"amlonllselJUluofleuulojul leUO!leN e Jo uofleluaLualdw! aql o1 alnqpluoo

pue 'ajfl 1o/flflenb eql aAoldw! 'q_44o1_OiLUOUOOa"S'FI alelnLUflS olsaiouaSe _aqlo pup VSVN /fq pau/eluieLu saseqelep Buisuas eloLua_

a81_1/uaA o41 jo 'laulelUl aql e!A 'esn o!lqnd peo_q alelnwfls ol "",

:s! leO8 uteJfioJdJeAO [tiE(2 aoual.o S aoeds pue ql.te._ 1o ash oflqnd Vlfl VSVN oq_L

;euJe;Ul eq; JeAO e_eGeoue!os, coedspue q;_e jo esF]--_lqnd

mergency

:$:8 :i:i:i::.

The system concept is to use remote sensing data set_ _ the NASA

DAACs, or other data providers, to produce new da£_ii_ers that areintegrated into a commercial Geographical Infi

These new data layers, when incorporated with the

infrastructure, and demographic data in the GIS, will

Alabama Emergency Management Agency with additional

for addressing flood related disasters.

GIS).

The second component of the system concept is to use the Internet as

primary communication path for transferring remote sensing data

data products from the data providers to the processing center, and

then to the Alabama Emergency Management Agency and its staff inthe field.

© SENTAR, Inc. 1997 97-0133

• Goal

Project'GOaland Objectives% ,% •

To develop a GIS-based, prototype system,for enhanced"_%_,....

emergency management of floods that: 1) utf_zes Remote SensingData (RSD), the Internet, and existing software

readily expandable into an all-hazards tool for use

Alabama or anywhere the necessary source data is

:ts and 2) is)ut

Objectives

- 1. Identify RSD, particularly from NASA's DAACs, that h_

potential to enhance flood management

- 2. Develop and demonstrate a prototype system for utilizing

for flood management by Alabama EMA staff

- 3. Demonstrate the value added of RSD for emergency

management operations

- 4. Demonstrate the feasibility of using the Internet as thecommunication network from the DAACs to end user.

© SENTAR, Inc. 1997 97-0133

Available'Data Sets•%.

Vector Data

TIGER-- 13 coverages from

TIGER 94 data covering 30Alabama counties

FEMA Q3 Floodplains-- 1%and 0.2% flood zones for 19

Alabama counties

Imagery derived water

boundary-- aera coverages

correspond to the Landsat TM

image areas

Four Landsat es

covering the

eastern portion of

• Images provided as

images including allbands, a simulated true

image, and a water

classification image

• Sample aerial photographyfrom USGS and SENTAR

photos

;outh

© SENTAR, Inc. 1997 97-0133

TIG ER ...........Data Coverages

CountiesNorthern

Blount

Colbert

Cullman

DeKalb

Etowah

Franklin

Jackson

Jefferson

Lauderdale

Lawrence

Limestone

Madison

Marshall

MorganSt.Clair

Walker

Winston

Southeastern

Barbour

Bullock

Coffee

CovingtonCrenshaw

Dale

Geneva

HenryHouston

Macon

MontgomeryPike

Russell

,%,...%

'%-+.

....." Data Coverag"-..... es

CensUs.TractfBlock Groups

Streets, ro_ds, highways•,k.::::::::::::.

Insterstates ...._;_iiiii!ii_>......

Waterways, rivers_d_er:water features

Railroads

State and county boundarie_

City and Township boundari_

School districts

104th Congressional districts

Voting districts

Apartment complexes

Trailor parks

Shopping centers

© SENTAR, Inc. 1997 97-0133 _iiii!iiiiii!iiiiiiiiiiiiiiiiiiiiiiiiiiii!!iiiiiiii_i

0

GIS Data Coverages for the Alabama EMATHEME

Hydrology

Utilities

Geology

Boundaries

FEMA Flood MapsHydrographyLake Levels

EAS Radio CoverageDoppler Radar (D)Bum Sites

Sewer Systems

Surface Geology (D)

Military InstallationsBuilt-Up AreasCounty BoundInternational Bound

Demographics (over 55)

USGS Control Points

-°_

'_'D_TA LAYERS

Wetlan_°'(D,)USACE Floo_ plainsStreamflows "_\

Fresh Water SystemS, .......

Fuel Facilities .......Water/Sewer ,,__Treatment Plants ...._

Elevation Contours

Snow & Rain Accumul. (DCoastal Hydrology

Socioeconomic

Geodetic

Hazards/Environmental

Man-Made Featuresand Structures

Photogrammetric

Cultural Resources

Land Use/Cover

Hazardous/Toxic/Radioactive Waste

Nuclear PlantsFire DepartmentsEOCsRoads & StreetsMaritime Ports

Photos and Imagery

Historical Standing Sites/Graveyards

Land Cover/Land Use

Evacuation RoutesEmergency PlanningEnhancements

Local Government BoundState Administrative BoundFederal Administrative BoundCensus Geography

Economic (income) (D)

Medical WastesHazardous Materials

BridgesAirportsHospitalsPolice DepartmentsDesignated Shelters

Transmission Lines (D)Electric Power grid (D)

_i_S.olidWaste Sites

Milit__ _.d.._Res tr ict"_'*_"i_ _! ii_!i_ii:_::_"_!_iii_:_

De mo grap hi_i!(_]_[_!

Disabled, Nursing Homes

' ii',iiiiiiiiiiiiiiiiiiiiiii 'ili! iiii@ I

PipelinesRailroadsPower PlantsGov't Man-MadeStructures

© SENTAR, Inc. 1997 97-0133

,!/

i:

I

i

!

E(D0_

09CD

00

nO-

f

iin

00

Ec_

:_!:::._]".._'..!_!:-::_;_: :i!__:_!!:_:_ _ _:!!!_!_! i!!:!:i!:_:::_:!:!_i:!ii:!!i:!!i:!!:!:i!i!!!!:!!!_:i!i!::!:_..':!_!i!i!!i!!!_:!ii:!i!!!!!:i:i:!ii:i:!:_:i:i:!:!:i:!!!i:i!i::i_i:!i__!:i _

•;_i_.._..::_._::_::_::_._::_:_._:_.:._:_:_i::_:_:_._:_._::_}:_:._::_:_::i:_;i_:_:i_:_:_:_.L:_._i_._:i:i_::_.:_._._.::::_:.:_:_._:_:_:_:_:_:]_:_:_::.:_:.:._

•_: .:: _:::'_::.: _::_:::::*:..,.;. _,

.__,.,,i_:_:_._ _'.

:::'_'_""':'":_;_ ....:::::::::::::::::::::::: .-:::i:_:::_!iiiiiiiii!iii!i_ !I

:::::::::::::::::::::::::i::::::i:i:f:!::i;i:_:_"

::[i]ii]]]?![]]!iiii!!ili¢/

.i!!iiiii!!i!ii_"

d

Z

@

System ........Prototype Concept

_'" ._.. RSD

& Requests '--- ,._

Spalial.I Fealure _

I _1 +_"?

I Spa_alDa_ I# _I sources

Sla_$pa'_alDa_

DalaProcessingCenter

1

Na bama

Sun Ser_r

Rood prediclon, warning, _. as_e_r_ntinxforrr_1¢n& images

• _!-

FEMA _._

Co un'_ I_1Ofioe

SPOT&o_e,_o /I

-/ I 01her RSD.t/ _r=aft I So.urce_

.,.--

Ai_ SPOT, GOES/ 'TVA, NWS,

NexRad, & O_erWea'her Si'les

+

/

/

• E-Mail ,5. JEM InformatonI,¢"E]'dData I Sources.

- -'-I_,RSeleot, d Warning,,o,s_ssrnent, _. I

Reoovery Images IGeneral PUblic8,Text J

Legend

PhysicalMedia

CellularCommunioatons

© SENTAR, Inc. I uu /

Inlemet

97-0133

Completed Tools and DataGIS data coverages for 17 data coverages for 17 northern

and 13 south eastern Alabama counties a_e been built;installed on AEMA computers and on CD

Four Landsat images processed and delivered

• USGS aerial photography for sample areas in AI_

processed and delivered on CD

EMO Tool user interface software program develo

installed on AEMA computers

Internet server scripts developed for use on Sun server

Tool and data assembled in Huntsville to process

additional remote sensing data.

© SENTAR, Inc. 1997 9%0133

What is Stillto Be done?

• Continue to build achive of remote sens_g!iidata.

• Integrate data products and tools into SOPs_i!! iiiiii_,_,:_,:,,,....,.,:::.:.:.:-:-:.:+:.:.:.:.:.:.:.:.:.:. :,=================================================

• Collect data and create data coverages for

state spatial data.

• Experiment with Civil Air Patrol images.

Integrate the Internet portions of the system at the AE

EOC (awaiting connection)

Expand coverage to remaining counties.

© SENTAR, Inc. 1997 97-0133

Emergency Management,Operations (EMO) Tool

• Tool Purpose - simpleuser interface

• Tool Setup for Files and

Programs

• Usage of Program Icons

• Usage of Menu Bar

• Usage of Windows in

Viewing Area

• Usage of Linked

Programs

• Tool Configuration

Flexibility

© SENTAR, Inc. 199797-0133

tt10-L6 Z66 t "aUl '_JVlNBS @

+:.::: >. :+ :.'+:.:.: -:i::::+:.:.:

:.,:: ::" :, ,:- :,::-::::.., .,::::::

i17i%ii:iiiiiiii:_i_,i:_:_:_.

• "<:.

i:.'.:x4-._:::'" :'_":::_i_:i&.: _

........ ::::::::::::::::::::" ".-'!!i:iiii:::::.... ':"'""

\ ..........::::::_::i..:_:i:::<i_!:•.:.:.:_.:.: : -.:.: : :.:

Post theNASA CAN Project

• The.,Intemet portion of the system

will b'eiifinalized and tested.

• The ges and the

'( .... remaining c .ties for the

, TIGER data b_

Remote sensing

will be processedavailable.

A home page on the

will be established to

and information to the

For project information on the Web:http ://iq uest. com/~ se ntar/NA SA/F Iood_M anageme nt.html

© SENTAR, Inc. 1997 97-0133

:......: .

_:_:_2_i_i_i_i_iiii:_iiiiiiii_:ii_iiiiiiiiiiiiiiii_iii_!i:ii_:_::!_!:i_::_L:_:i::_:_:_:_:::_::::_:_:............. ......